基于信任模型的机器人共享控制研究

发布时间：2018-03-16 22:15

本文选题：共享控制　切入点：主从遥操作　出处：《吉林大学》2017年硕士论文　论文类型：学位论文

【摘要】：人类对自然资源开发和利用的过程中,机器人发挥的作用越来越大。以现有的控制和人工智能水平,短期内研制出在未知环境下全自动机器人是一个难以达到的目标,将机器人自主控制和操作者的高级决策融合在一起是目前的研究趋势。但是在现有的几种共享控制方式中,操作者控制和机器自主控制分别控制机器人的不同自由度,或者在时间上予以划分,人类决策和机器人智能在控制回路上依然相对独立,人机融合度不高。为了解决共享控制中的人机融合问题,本文建立了基于信任模型的机器人共享控制策略。首先根据人工势场理论建立基于视觉的虚拟引导力,然后通过基于人机信任模型的模糊逻辑系统实时得到机器自主控制和操作者控制所占的权重,最后采用力融合机器人共享控制方法,将人手力和虚拟引导力融合,由位置预测器得到运动趋势,控制手控器的运动和机器人的位置随动,从而实现人类智能与机器智能的集成。针对基于信任模型的共享控制策略,本文主要完成了以下工作:1、分别计算手控器和机器人的运动学,设计手控器坐标系到机器人坐标系的位置和姿态的映射关系。2、建立人机信任模型,详细说明了操作者、机器人、环境等影响对人机信任关系的影响。在遥操作系统运行过程中,人机之间的信任水平始终保持在一个合适的水平。3、将人机信任度和机器人的运行状态这两个量作为输入,根据信息融合的模糊逻辑法,建立模糊逻辑系统,由模糊推理得到操作者控制和机器人自主控制各自实时的权重。4、建立基于信任模型的共享控制策略,操作者的操作力和机器人自主引导力分别乘以相应的权值系数并相加,使用融合力控制手控器和机器人运动。5、在VS+Qt环境下编写控制程序,采用多线程编程的方式,并以合理的时序安排各子功能模块的运行。在搭建好的实验平台上进行实验,验证本文共享控制策略的有效性。
[Abstract]:In the process of human exploitation and utilization of natural resources, robots play a more and more important role. With the existing level of control and artificial intelligence, it is difficult to achieve the goal of developing fully automatic robots in the unknown environment in the short term. It is a research trend to combine robot autonomous control with advanced decision making. However, among the existing shared control methods, operator control and machine autonomous control respectively control the different degrees of freedom of the robot. Or divided in time, human decision-making and robot intelligence are still relatively independent in the control loop, the degree of human-computer fusion is not high. In order to solve the problem of human-computer fusion in shared control, In this paper, the shared control strategy of robot based on trust model is established. Firstly, based on the theory of artificial potential field, the virtual guiding force based on vision is established. Then, the weights of machine autonomous control and operator control are obtained by fuzzy logic system based on man-machine trust model in real time. Finally, the force fusion robot sharing control method is used to combine the manual force and virtual guiding force. The motion trend is obtained from the position predictor, and the movement of the hand controller and the position follower of the robot are controlled, so that the integration of human intelligence and machine intelligence is realized, and the shared control strategy based on trust model is proposed. This paper mainly completes the following work: 1, calculates the kinematics of the hand controller and the robot, designs the mapping relationship between the position and attitude of the hand controller coordinate system to the robot coordinate system, establishes the man-machine trust model, explains the operator, the robot, the robot, the robot, the robot, the robot, the robot, the robot, the robot, the robot and the robot. The influence of environment on man-machine trust. During the operation of remote operating system, the level of trust between human and computer is always kept at a suitable level .3. the two variables of man-machine trust and the running state of robot are taken as input. According to the fuzzy logic method of information fusion, the fuzzy logic system is established, and the real-time weights of operator control and robot autonomous control are obtained by fuzzy reasoning, and the shared control strategy based on trust model is established. The operator's operating force and the robot's autonomous guiding force are multiplied by the corresponding weight coefficients and added together respectively, and the fusion force is used to control the hand controller and the robot's movement .5. the control program is written under the environment of vs QT, and the multithread programming method is adopted. The operation of each sub-function module is arranged with reasonable timing, and the experiment is carried out on a good experimental platform to verify the effectiveness of the shared control strategy in this paper.
【学位授予单位】：吉林大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TP242

【相似文献】